Forming electrochemically active biofilms from garden compostunder chronoamperometry

Sandrine Parot, Marie-Line Delia, Alain Bergel *

Laboratoire de Genie Chimique-CNRS-INPT, 5 Rue Paulin Talabot, 31106 Toulouse, France

Abstract

Dimensionally stable anodes (DSA) were polarized at different constant potential values for several days in garden compost. After aninitial lag period ranging from 1 to 10.5 days, the current increased fast and then stabilized for days. Current densities higher than100 mA m�2 and up to 385 mA m�2 were obtained with the sole organic matter contained in compost as substrate. Control experimentsperformed with sterilized compost, oscillations of the current with the temperature, kinetics of the exponential phase of current increaseand observations of the surface of electrodes by epifluorescence microscopy showed that the current was controlled by the colonization ofthe electrode surface by a biofilm which originated the indigenous flora of compost. Three individually addressed electrodes polarized atdifferent potentials in the same reactor led to identical current evolutions on each electrode, which underlined the key role of the micro-bial flora of the compost in the discrepancy observed in the other experiments. Chronoamperometry revealed a promising technique tocheck natural environments for new electrochemically active microbial species.

Keywords: Electrochemically active biofilm; Biofilm electrochemistry; Compost; Dimensionally stable anode; Chronoamperometry

1. Introduction

It has been recently demonstrated that several micro-organisms have the ability to form efficient electro-catalyticbiofilms on electrode surfaces. In several cases the electro-catalytic properties of biofilms have been clearly relatedto the presence of some strains that are able to exchangeelectrons with the electrode either directly (Geobacter sul-

furreducens, Rhodoferax ferrireducens, etc.) or through nat-ural electron mediators (Shewanella spp., Geothrix

fermentans, etc.) (Lovley, 2006). Nevertheless, most studiesdeal with complex microbial consortia, which are imple-mented in microbial fuel cells (MFCs). MFCs have beenset-up in natural sites such as marine or freshwater sedi-ments (Holmes et al., 2004; Tender et al., 2002) or in labo-ratories with inoculums coming from a wide range of

different microbially-rich environments: anaerobic sewagesludge (He et al., 2005; Jong et al., 2006), activated sludge(Kim et al., 2004; Rabaey et al., 2004), industrial anddomestic effluents (Heilmann and Logan, 2006; Logan,2005), and animal waste (Min et al., 2005). At the momentnatural soils have not been studied as a potential source ofelectrochemically active microorganisms yet, but it hasbeen reported that heat treated soils could be used success-fully as a source of spore forming bacteria that producehydrogen (Niessen et al., 2006; Rosenbaum et al., 2006).

The purpose of this work was to propose a chronoampe-rometry procedure suited for identifying electrochemicallyactive biofilms in natural environments. Until now chrono-amperometry has been implemented only very rarely toinvestigate natural environments, although it should avoidany disturbance and artefact due to the second electrode orthe internal resistance of the set-up, as may be the case withmicrobial fuel cells. As microbial biofilms are involved inmany biogeochemical cycling of metals (Fe, Mn, and tracemetals) it was expected that compost would constitute a

* Corresponding author. Tel.: +33 (0) 5 34 61 52 48; fax: +33 (0) 5 34 6152 53.

E-mail address: Alain.Bergel@ensiacet.fr (A. Bergel).

promising environment for investigation. Moreover a pre-liminary study by our group has suggested that gardencompost may be a pertinent environment in which to findelectrochemically active biofilms (Dulon et al., 2007).Experiments were run under well-controlled electrochemi-cal conditions, by imposing a constant anodic potentialthrough a traditional three electrodes system. Most pre-vious studies have been carried out with graphite and car-bon anodes, with differing morphologies (cloths, foams,etc.) and sometimes modified with catalytic compounds.Here, dimensionally stable anode (DSA) electrodes wereused. This kind of electrode made of titanium covered byiridium and tantalum oxides has been designed to ensurehigh electro-catalytic activity (Trasatti, 2000), and theyhave proved very efficient in many industrial oxidation pro-cesses (Comninellis and Nerini, 1995; de Andrade et al.,1998; Menini et al., 2005). Because of their rough surfaceand their optimal electro-catalytic properties it wasexpected that DSA would be an efficient tool to detectnew electrochemically active biofilms.

2. Methods

2.1. Chemicals and materials

The compost for biological cultivation (Eco-Terre) waspurchased from a garden centre. 2 L of a sodium chlorideNaCl 10 mM (S-9625, Sigma) solution were mixed withcompost to obtain a 3 L final working volume. The tem-perature in the reactor was measured with a PT100 temper-ature sensor (Sofraico), the oxygen concentration with aMettler-Toledo Ingold InPro 6800 O2 sensor (detectionlimit 6 ppb) and pH with a Radiometer AnalyticalPHM210 standard pH Meter. All measurements were car-ried out at room temperature. Control experiments wereperformed in compost sterilized by 20-min autoclaving at121 �C.

2.2. Electrochemical measurements

A conventional three-electrode system was implementedwith a multi-potentiostat (VMP2 Bio-Logic SA) interfacedto a computer (software EC-Lab v.8.3, Bio-Logic SA). Theworking electrode was a 10 · 2.5 · 0.1 cm dimensionallystable anode electrode (DSA�, Electro Chemical Service)with a 20 cm2 working surface area. 8 cm of the DSA wereembedded vertically in the compost and 2 cm were abovethe level of the compost and connected with a flat alligatorclip. Before each experiment, the electrode was cleaned by5-h galvanostatic electrolysis at 20 mA cm�2 in a 0.1 M sul-phuric acid solution. The counter electrode was a 35 cm2

surface area graphite rod (Alfa Aesar, stock#10134). Allpotentials were monitored against a saturated calomelstandard reference electrode (Radiometer Analytical,TR100) protected with a second porous frit. For experi-ments carried out with three DSA electrodes in the samecompost reactor, a system was used which made it possible

to individually address three working electrodes withrespect to one reference electrode and one counter elec-trode (Bio-Logic SA).

For control experiments, electrodes were cleaned withethanol and rinsed with distilled water.

3. Results

3.1. Current generation in compost

The general procedure used here for the formation anddetection of electrochemically active biofilms consisted inchronoamperometry at constant potential. A constantpotential value was imposed for days to DSA electrodes(surface area 20 cm2) dipped in compost mixed with a10 mM NaCl solution and the current evolution was mon-itored as a function of time. Twelve experiments werespread over a period of one year with different batches ofthe same brand of compost (Table 1). Chronoamperome-tries were performed either with only one working elec-trode in 3 L reactors (experiments I–IV and VII–XII inTable 1), or with working electrodes set-up in 10 L reactorsand electrically individually addressed (experiments V–VIIin Table 1). Fig. 1 shows four chronoamperometries char-acterizing the current density increase (experiments I–IV inTable 1). The typical pattern of the current evolutionremained unchanged whatever the potential value, with 3successive steps: a lag period for around 5 days, a rapidincrease and a stabilization phase, which could go on for10 days. Five parameters were listed to describe the currentdensity evolution (Table 1). The current density was con-sidered to start to increase when it rose to more than1 mA m�2 in 1 day (Dstart in column 3). Column 4 givesthe maximal current density reached and the correspondingdate. The average value of current density at final plateauand the total duration of the stabilization period arereported in column 5. The parameter Dstart varied withina range of 1–10.5 days, but all experiments led to signifi-cant current increase. Maximal current density was onaverage around 105 mA m�2 for the 10 electrodes polarizedat 0.50 V/SCE, and it went up to 385 mA m�2 (experimentVIII). The stabilization phase presented quite different fea-tures. In some cases current remained close to the maximalvalue for days, while it sometimes decreased continuouslyfrom the maximal value (experiment VI). The variation inthe compost batches and in operating conditions certainlyexplains these discrepancies. It is important to note that 3the electrodes dipped in a single reactor and independentlypolarized at 0.50 V/SCE (experiment V) showed exactly thesame evolution and gave equal values of current and Dstart,showing that the environmental conditions (composition ofthe compost, organic matter available, indigenous flora,temperature and hygrometry during the experiments, etc.)controlled the current increase to a large extent.

The pH of the environment was initially 7.7 anddropped to 7.4 after the current increase. After 1 day, thevalue of the concentration of oxygen in the reactor was

inferior to the detection threshold of the sensor (6 ppb).The compost was hence considered as an anaerobic envi-ronment. At the end of current increase, epifluorescentmicroscopy images showed that the electrode surface wasfully covered by a biofilm composed by bacteria whichformed colonies spread all over the electrode (data notshown). A temperature sensor put in the compost in experi-ment VI revealed a complete correspondence between thetemperature oscillations and those of the current density(Fig. 2). The oscillations that were often observed duringthe current density increase (Fig. 1 for instance) were there-fore due to the circadian temperature variations.

Four independent control experiments were performedat 0.10, 0.40, 0.50, 0.70 V/SCE (Fig. 3) with compost which

was preliminary sterilized by autoclaving. A low reductioncurrent was observed at 0.10 V/SCE while oxidationoccurred at more positive potential values (inset inFig. 3), but the current density dropped close to zero intwo days. No current was then obtained even after 10 dayswhatever the polarization. The presence of the indigenousflora was consequently required in order that the electro-chemical oxidation occurred.

3.2. Influence of polarization

In order to avoid the discrepancies due to the differentcompost batches, the influence of polarization on the cur-rent increase was studied with three working electrodes

Table 1Current density characteristics determined from 12 chronoamperometries achieved with DSA electrodes embedded in compost (NaCl 10 mM added) at0.10, 0.30, 0.40, 0.50 and 0.70 V/SCE

Experiments(roman numerals)and electrodes

Polarization(V/SCE)

Start of current increaseDstart (day)

Maximum current density(mA m�2); date of maximalcurrent (day)

Average value at stable state(mA m�2); duration ofstabilization (days)

I 0.10 5 51; 17 49; 10II 0.40 6 24; 12.5 21; 7III 0.50 4 62; 19 52; 6.5IV 0.70 5 70; 17 66; 4V Electrode 1 0.50 3 105; 9 67; 4.5

Electrode 2 0.50 3 106; 9 68; 4.5Electrode 3 0.50 3 106; 9 68; 4.5

VI Electrode 1 0.10 4 17; 13 No stabilizationElectrode 2 0.50 4 19; 13 No stabilizationElectrode 3 0.50 4 19; 13 No stabilization

VII Electrode 1 0.30 10 145; 13.5 129; 2Electrode 2 0.50 10 128; 13.5 114; 2Electrode 3 0.70 10 92; 15.5 88; 2

VIII 0.50 4 385; 9 229; 9IX 0.50 1 53; 3 36; 2.5X 0.70 No significant increaseXI 0.50 5 62; 10 59; 2XII 0.70 8 76; 18 Electrode removed before stabilization

Experiments V–VII were performed with 3 electrodes in the same reactor.

Fig. 1. Current density plotted against time for 4 DSA electrodes polarized at 0.10; 0.40; 0.50 and 0.70 V/SCE (curves a–d, respectively) in 4 differentreactors containing compost and NaCl 10 mM (experiments I–IV).

individually addressed in a single 10 L reactor (experimentVII) and polarized at 0.70, 0.50 and 0.30 V/SCE. For thethree electrodes there was an identical initial lag phase of10 days and then the current increased from the 10th tothe 13th day. Stabilization occurred during the last twodays (Fig. 4). The general shape of the curves was similarfor the three electrodes. The potential value actually hadonly a significant effect on the final current density values:on average 129, 114 and 88 mA m�2 at the stable plateaufor 0.30, 0.50 and 0.70 V/SCE respectively (these average

values are different from the values given in Table 1 thatreports the maximum values).

As seen in Table 1, the polarization performed at0.50 V/SCE provided the most reproducible values of cur-rent density, whereas polarizations at 0.70 V/SCE did notsupply higher currents or were sometimes found to producelower or slower current increase. Experiments IX and Xwere run in parallel at the same time with the same batchof compost, but in two independent reactors. In experimentX the DSA polarized first at 0.70 V/SCE during 8 days

Fig. 2. Temperature and current density evolution plotted against time in a reactor containing compost and a DSA electrode polarized at 0.50 V/SCE(experiment VI). The Y-axis on the left represents the current density in mA m�2 (curve down) and the Y-axis on the right represents the temperature inCelsius degrees (curve up).

Fig. 3. Current density evolution plotted against time with a DSA electrode polarized at 0.50 V/SCE embedded in sterilized compost. Inset gives thebeginning of the control experiments, before current density reached 0 mA m�2, for polarization values of 0.10 V/SCE (curve a), 0.40 V/SCE (curve b),and 0.70 V/SCE (curve c).

gave a weak current density (Fig. 5). Then, when the poten-tial value was switched to 0.50 V/SCE (arrow in Fig. 5), thecurrent density increased for more than 8 days and reached60 mA m�2 only 15 days after the polarization change. Onthe contrary, in experiment IX the electrode initially polar-ized at 0.50 V/SCE showed a fast current increase from thefirst day to around 50 mA m�2 on day 3. This confirmedthat the potential value of 0.50 V/SCE favoured the forma-tion of electrochemically active biofilm more consistentlythan higher values of potential.

4. Discussion

The general trend of current evolution recorded withDSA electrodes polarized in compost was coherent withthe four-phase pattern growth that is characteristic of the

formation of a bacterial biofilm (Costerton et al., 1999;Watnick and Kolter, 2000). The initial lag phase of1–10.5 days (Table 1) was consistent with the first phaseof bacteria approach and attachment to the electrode sur-face. During this period, the bacteria were adapting tothe sessile conditions at the surface and to the local physi-cochemical conditions induced by polarization. The com-pactness of the medium and certainly local heterogeneityin composition and bacterial contents may explain the dif-ference in length of this step. In microbial fuel cells, biofilmformation has often been reported to be longer, between 6and 15 days (Chaudhuri and Lovley, 2003; Jang et al.,2004; Liu et al., 2005), therefore the compost may be thesource of microorganisms that are able to adapt pretty fastto the anode conditions. A transient peak often appearedduring the initial lag phase (easily visible in Fig. 1 between

Fig. 4. Current density plotted against time from current recorded with 3 individually addressed DSA electrodes dipped in the same compost reactor.Curves 1–3: imposed potential values were 0.30; 0.50 and 0.70 V/SCE, respectively (experiment VII).

Fig. 5. Effect of polarization on current density evolution. Curve 1 (experiment X) the imposed potential initially set to 0.70 V/SCE was switched to0.50 V/SCE at day 7 (arrow). Curve 2 (experiment IX) the imposed potential was 0.50 V/SCE from the beginning. Both experiments were performedconcurrently in two different reactors.

days 1 and 4, or in Fig. 4 before day 2, for instance) whenthe compost turned anaerobic. This peak might be there-fore due to the oxidation of some compounds producedby the anaerobic bacterial metabolism, in the same mannerthat sulphide compounds resulting from the sulphatereducing bacteria metabolism are oxidized in sediments.

The current increase that followed the lag period cor-responded to the active growth phase of the biofilm anda stationary phase occurred when the surface coveragewas maximal or the nutrient supply became limiting. Theexperiments did not last long enough to enter the fourthphase of biofilm decline, but the final slow current decreasewas more likely due to substrate depletion. For instance,the current decrease of a few tenths of milliamps that wasgenerally observed during one day just after reaching themaximal value may express the slow depletion of substratein the vicinity of the electrode surface (Figs. 4 and 5). It isconsistent with what was previously observed when theaddition of acetate (10 mM) to compost as a supplemen-tary nutrient caused a rapid rise of the current density(Dulon et al., 2007).

The microbial origin of the catalysis was confirmed bythe control experiments performed after compost autoclav-ing, which did not result in current increase whatever thepolarization value. The weak current observed at the begin-ning of the polarization, which decreased quickly (Fig. 3),may be due to the formation by autoclaving of a com-pound that directly reduced on the DSA at 0.10 V/SCE(negative current, see curve a in inset of Fig. 3) and oxi-dized at higher potential values (positive current, see curvesb and c in inset of Fig. 3). This was similar to what has pre-viously been observed in autoclaved aquatic sediments withgraphite anodes where current has been measured duringthe first hours of polarization (Holmes et al., 2004).

The small circadian oscillations of the current (Fig. 2)have also been observed with electrochemically active bio-films formed from drinking water on platinum micro-elec-trodes (Dulon et al., 2007). It perfectly correlated with thehigh influence of the temperature on microbial metabo-lisms. All these observations confirmed that the electro-chemical process was controlled by the microbial activityof the biofilms formed from the indigenous flora of com-post. The biofilm revealed itself able to oxidize organiccompounds present in compost and to transfer the elec-trons to the DSA electrode.

Considering all experiments, it appeared that polariza-tion at 0.70 V/SCE gave smaller currents (experimentVII) or longer initial lag phase (experiment X). It may bespeculative to compare other experiments that were notachieved simultaneously and with exactly the same com-post batch, as the dispersion due to environmental condi-tions does not allow to drawn firm conclusions on theeffect of potential. Nevertheless, the results obtained at0.70 V/SCE were never significantly better than at 0.50V/SCE and appeared less reproducible. Optimal conditionsto be sure to form an electrochemically active biofilm werehence to impose a potential around 0.50 V/SCE. Similarly,

it has already been demonstrated that the polarization ofgold electrodes influenced the morphology and the biofilmstructure of Pseudomonas fluorescens which developed ontheir surface (Busalmen and de Sanchez, 2005). In this casethere was no electron transfer between electrode and bacte-ria, but it was observed that the biofilm developed at 0.50 Vvs. Ag/AgCl following the same phases as observed here incompost: a lag phase during 30–40 min followed by expo-nential growth. On the contrary, polarization at 0.80 Vvs. Ag/AgCl stopped the biofilm development.

For further considerations on polarization influence, itwas supposed that the current I is proportional to the bio-film growth. In these conditions the traditional microbialgrowth equation:

dX=dt ¼ lX

where X represents the biomass concentration and l thespecific growth rate, can be expressed through the currentI:

dI=dt ¼ lI

The specific growth rate l (d�1) was calculated from thecurrent obtained with the 3 electrodes polarized at 0.30,0.50 and 0.70 V/SCE in the same reactor in order to avoidthe discrepancies due to the different compost batches(experiment VII). Plotting the logarithm of current densityagainst time for the increasing phase (starting around day10 for 0.30 and 0.50 V/SCE and day 11 for 0.70 V/SCE)gave l = 1.2 d�1 (straight line a in Fig. 6) for the potentialvalues 0.30 and 0.50 V/SCE, while polarization at 0.70V/SCE led to l = 0.8 d�1. These values are fully coherentwith traditional bacterial growth rates (Bailey and Ollis,1986). Electrochemically active biofilms formed on graph-ite anodes polarized at 0.20 V vs. Ag/AgCl from pure cul-ture of Geobacter sulfurreducens have also shown specificgrowth rate around 0.96 d�1 during their exponentialincreasing phase (Bond and Lovley, 2003). As observedin traditional microbial cultures, the curve derived fromthe linear regression due to nutrient limitation. A commonvalue l = 0.4 d�1 fitted this nutrient-deprived phase, what-ever the potential value imposed to the electrode (straightline b in Fig. 6). It may be concluded that the growth phaseof the electrochemically active biofilms revealed to be fullycontrolled by biochemical parameters and to be indepen-dent of the potential as long as too high value (0.70V/SCE) did not interfere in its formation.

The current densities often higher than 100 mA m�2 andup to 385 mA m�2 obtained without any addition of sub-strate in the compost were among the highest valuesreported previously for compact media. Current densityof 34 mA m�2 (Holmes et al., 2004) have been obtainedwith three different sediment types in laboratory MFC,and values around 40 mA m�2 (Ryckelynck et al., 2005)and 100 mA m�2 (Tender et al., 2002) have been reachedwith graphite electrodes embedded in situ in marine sedi-ments. Obviously, higher current densities have beenobtained in liquid media. Chronoamperometry performed

with graphite electrodes in pure cultures of microorganismsisolated from sediments have given a large range of resultssuch as 165 mA m�2 with Desulfuromonas acetoxidans,268 mA m�2 with Geobacter metallireducens (Bond et al.,2002) and even 1143 mA m�2 with Geobacter sulfurredu-cens (Bond and Lovley, 2003). Actually, it seems that someof the highest current densities, higher than 16 A m�2

(Niessen et al., 2006) or 30 A m�2 (Rosenbaum et al.,2006), have been achieved with spore forming bacteriacoming from heat treated compost. Heat treatment was away to select hydrogen producing microorganisms thatwere then used in solution. The hydrogen produced in solu-tion was oxidized abiotically on platinum or tungsten car-bide anodes. The principle of the fuel cell was thereforebasically different from work presented here, as it did notimplement a direct biofilm-driven catalysis.

5. Conclusion

This study confirms that soils are an interesting andeasy-to-handle source of electrochemically active micro-organisms which is ‘‘available everywhere, free of charge’’(Niessen et al., 2006). Work is now in progress to identifyand isolate the predominant bacterial species that formthe most efficient biofilms. It should be noted that noneof the 18 chronoamperometry experiments failed in form-ing electrochemically active biofilms. Chronoamperometrycan be considered as a promising tool to drawn new elec-trochemically active biofilms from all kind of natural andindustrial environments that has not been checked yet.

Acknowledgements

The authors are most grateful for technical contribu-tions from Luc Etcheverry, engineer CNRS, Laboratoirede Genie Chimique (Toulouse, France), and for the kind

advice from Pr. Andre Savall and Dr. Karine Groenen-Serrano from Laboratoire de Genie Chimique, UniversitePaul Sabatier (Toulouse, France). They deeply thank Dr.Alfonso Mollica from CNR-ISMAR (Genoa, Italy) forthe design of the electrochemical set-up and for his efficientadvice. This work was part of the EA-Biofilms project sup-ported by European Community (FP6-NEST508866).

References

Bailey, J.E., Ollis, D.F., 1986. Biochemical Engineering Fundamentals,second ed. McGraw-Hill, New York.

Bond, D.R., Holmes, D.E., Tender, L.M., Lovley, D.R., 2002. Electrode-reducing microorganisms that harvest energy from marine sediments.Science 295, 483–485.

Bond, D.R., Lovley, D.R., 2003. Electricity production by Geobacter

sulfurreducens attached to electrodes. Appl. Environ. Microbiol. 69,1548–1555.

Busalmen, J.P., de Sanchez, S.R., 2005. Electrochemical polarisation-induced changes in the growth of individual cells and biofilms ofPseudomonas fluorescens (ATCC 17552). Appl. Environ. Microbiol.71, 6235–6240.

Chaudhuri, S.K., Lovley, D.R., 2003. Electricity generation by directoxidation of glucose in mediatorless microbial fuel cells. Nat.Biotechnol. 21, 1229–1232.

Comninellis, Ch., Nerini, A., 1995. Anodic oxidation of phenol in thepresence of NaCl for wastewater treatment. J. Appl. Electrochem. 25,23–28.

Costerton, J.W., Stewart, P.S., Greenberg, E.P., 1999. Bacterial biofilms: acommon cause of persistent infections. Science 284, 1318–1322.

de Andrade, A.R., Donate, P.M., Alves, P.P.D., Fidellis, C.H.V., Boodts,J.F.C., 1998. Ethanol electro-oxidation in ruthenium-oxide-coatedtitanium electrodes. J. Electrochem. Soc. 145, 3839–3843.

Dulon, S., Parot, S., Delia, M-L., Bergel, A., 2007. Electroactive biofilms:new means for electrochemistry. J. Appl. Electrochem. 37, 173–179.

He, Z., Minteer, S.D., Angenent, L.T., 2005. Electricity generation fromartificial wastewater using an upflow microbial fuel cell. Environ. Sci.Technol. 39, 5262–5267.

Heilmann, J., Logan, B.E., 2006. Production of electricity from proteinsusing a microbial fuel cell. Water Environ. Res. 78, 531–537.

Holmes, D.E., Bond, D.R., O’Neil, R.A., Reimers, C.E., Tender, L.R.,Lovley, D.R., 2004. Microbial communities associated with electrodes

Fig. 6. Calculation of the specific growth rate coefficient l for 3 DSA electrodes in the same reactor (experiment VII) and polarized at curve 1: 0.30V/SCE; curve 2: 0.50 V/SCE; curve 3: 0.70 V/SCE. Dotted straight lines indicate the linear regression curves.

harvesting electricity from a variety of aquatic sediments. Microb.Ecol. 48, 178–190.

Jang, J.K., Pham, T.H., Chang, I.S., Kang, K.H., Moon, H., Cho, K.S.,Kim, B.H., 2004. Construction and operation of a novel mediator-and membrane-less microbial fuel cell. Process Biochem. 39, 1007–1012.

Jong, B.C., Kim, B.H., Chang, I.S., Liew, P.W.Y., Choo, Y.F., Kang,G.S., 2006. Enrichment, performance, and microbial diversity of athermophilic mediatorless microbial fuel cell. Environ. Sci. Technol.40, 6449–6454.

Kim, B.H., Park, H.S., Kim, H.J., Kim, G.T., Chang, I.S., Lee, J., Phung,N.T., 2004. Enrichment of microbial community generating electricityusing a fuel-cell-type electrochemical cell. Appl. Microbiol. Biotechnol.63, 672–681.

Liu, H., Cheng, S., Logan, B.E., 2005. Production of electricity fromacetate or butyrate using a single-chamber microbial fuel cell. Environ.Sci. Technol. 39, 658–662.

Logan, B.E., 2005. Simultaneous wastewater treatment and biologicalelectricity generation. Water Sci. Technol. 52, 31–37.

Lovley, D.R., 2006. Bug juice: harvesting electricity with microorganisms.Nat. Rev. Microbiol. 4, 497–508.

Menini, R., Henuset, Y.M., Fournier, J., 2005. Electrochemical oxidationof phenolic compounds using a flow-through electrolyser with poroussolid electrodes. J. Appl. Electrochem. 35, 625–631.

Min, B., Kim, J.R., Oh, S.E., Regan, J.M., Logan, B.E., 2005. Electricitygeneration from swine wastewater using microbial fuel cells. WaterRes. 39, 4961–4968.

Niessen, J., Harnisch, F., Rosenbaum, M., Schroder, U., Scholz, F., 2006.Heat treated soil as convenient and versatile source of bacterialcommunities for microbial electricity generation. Electrochem. Com-mun. 8, 869–873.

Rabaey, K., Boon, N., Siciliano, S.D., Verhaege, M., Verstraete, W., 2004.Biofuel cells select for microbial consortia that self-mediate electrontransfer. Appl. Environ. Microbiol. 70, 5373–5382.

Rosenbaum, M., Zhao, F., Schroder, U., Scholz, F., 2006. Interfacingelectrocatalysis and biocatalysis using tungsten carbide: a high perfor-mance noble-metal-free microbial fuel cell. Angew. Chem. 118, 1–4.

Ryckelynck, N., Stecher III, H.A., Reimers, C.E., 2005. Understandingthe anodic mechanism of a seafloor fuel cell: interactions betweengeochemistry and microbial activity. Biogeochemistry 76, 113–139.

Tender, L.M., Reimers, C.E., Stecher III, H.A., Holmes, D.E., Bond,D.R., Lowy, D.A., Pinobello, K., Fertig, S.J., Lovley, D.R., 2002.Harnessing microbially generated power on the seafloor. Nat.Biotechnol. 20, 821–825.

Trasatti, S., 2000. Electrocatalysis: understanding the success of DSA�.Electrochim. Acta 45, 2377–2385.

Watnick, P., Kolter, R., 2000. Biofilm, city of microbes. J. Bacteriol. 182,2675–2679.